Posts by karlp
36 posts • joined Thursday 11th February 2010 21:54 GMT
We already know, 100$ cheaper than the last one.
Karl P
Exit Strategies
I am sure that one or two of these companies could, by showing some initiative, take their experience working with accelerating certain workloads and turn it more into a consultancy gig where they help companies with general app acceleration and workload efficiency tuning.
That said, most of the these flash-repackagers+secret sauce would do well to start shopping their engineering talent to the same companies they have been buying the underlying dies from.
Flash is growing up awfully quick right now, hopefully most of these people can avoid being caught-out.
Karl P
An Important Consideration
I sometime play in this arena, and I want to add a few things before everyone piles on.
It's important to note that while, from what we read here, the project may seem to have been a bit overambitious, there is some precedent for this being a worthwhile, albeit expensive project.
There have been a few networks and large, geographically dispersed, production houses who have stepped up to get off of tape, get all of their assets digitized (within reason) and provide access to their data in practically all of their edit-bays and injest-points.
This is a hugely expensive thing to do. Think PB level's of SAN, MAID, metadata and asset management systems. Now think about linking that out at 4Gb or better speed to many dozens or even hundreds of geographically dispersed points.
It's a complicated project, but one that has been done, and has delivered substantial productivity benefits and even many cost benefits. (You would be quite shocked the price of a tape-based infrastructure. Everything from 50$ tapes that won't be reused to 50'000$ tape-decks that need considerable ongoing maintenance.)
Whether or not everyone needs access from everywhere is highly debatable, but by some measures it wouldn't be that difficult. All you need is a server hanging off the san, tied to the metadata and asset management databases with a web front-end and some custom software rendering the video down to h.264 streams in real time. Maybe use some heavy CUDA or something. Once you had the rest of the system functioning, this might be a not-that-big-of-a-deal bolt-on.
I'd be curious to know where they stand today. The project, from what I can surmise so far, is actually not only valid, but probably still needs to be done, albeit under much different direction.
Karl P
Pricing
I would be shocked if even a basic two node config and 6 cheap spindles is less than 15K.
I would anticipate a mid 20's price, for basic normal configs, and maybe 30K with full compute nodes. Even then I am very interested.
The addition of a high speed shared storage subsystem could be a clincher here. I'm guessing this will be almost DAS performance in a shared subsystem, something you would need to spend big money on to get today in a true external shared storage system. The fact it will be limited to four compute nodes is acceptable, considering the workload I can accomplish with 8 sockets these days.
I love it.
While I do have concerns about chassis reliability and hopes for the availability of a spares kit, the raw idea I love. I hope they do well.
Re: In an ideal world...
I install locks on my doors.
They exist for no other reason than to keep the honest people out and enable me to get homeowners insurance.
They are also by many accounts an inconvenience; trying as I am to juggle for this and that while attempting to get the insufferable key to mate with the door.
Judging by the fact that I have been burgled before, I can say with some confidence that they do nothing against people who want to get in.
DRM is similar. It gives honest people pause and provides the basis for which content providers feel like some effort has been made.
It really isn't a lot more complicated than that.
I do still wish I lived in a world which didn't need locks, but that is a different issue altogether.
Re: What a huge fraud! Adobe worse than Electronic Arts (SimCity fraud)...
I am in the same boat as you, I have a client who is now licensed up on 4 or 5 (maybe more now) copies of creative cloud. They were moving off of one legit CS5 and a bunch of bootlegs. They couldn't justify frontloading 15'000$+ of adobe software. 200~300$ish a month is seen as an incredible bargain. I wouldn't be surprised if they end up with 10 or 15 seats when it is all said and done.
You guys gotta remember that adobe has never been the right company to do business with if you aren't in the business of making money with your tools. But if you are in that business - the subscription options are perfectly reasonable.
Because it has come up as a parallel, I also have a number of clients getting ready to go office 365 E3. The price hurts a bit, but nowhere near as much as front loading 500$+ per Office OpenLicense. Plus they get to get rid of their exchange servers (or google apps accounts) and their skype premium group-conference accounts (for lync). This is to say nothing of the support niceties you gain.
In most cases when we are doing an ROI on SaaS, we are seeing payback times at or over 3 years vs buying it ourselves. It would be stupid not to seriously consider that.
I understand that for home users there are different realities, and those may need to be addressed in better ways and/or at more flexible price points.
Do try and remember, for as many people as are here complaining about the SaaS movement, there are many, many more that are either indifferent or even in support of these changes.
However....
While the jury is still out on the general concept of leased software, the article and some of the comments are a bit off.
They aren't taking away the ability to purchase the current version of office (2011).
I am sure if you really need 2008, you can continue to get that under volume licensing as per usual, or ebay.
While it has some new (and some old) bugs still about, Office 2011 is _vastly_ faster, at least in my experience (100+ nodes of various apple machines).
On the subject of 365 - it gets you access to install more than one copy if your needs fall within the licensing TOS.
Additionally, it gets you outlook, which makes the comparison more like 200$+ vs 100$ a year. Finally, you do get access to one note and other services, albeit in their online versions.
While I am far from a Microsoft evangelist, there are plenty of things to be upset about without making up new ones.
Karl P
Re: Technical Matter
Indeed NASA are aware and in support of the matter.
In Jeff's Bezos blog on the matter (already linked in this conversation) he states the following:
"
Finally, I want to thank NASA. They extended every courtesy and every helping hand – all of NASA’s interactions were characterized by plain old common sense, something which we all know is impressive and uncommon. We're excited to be bringing a couple of your F-1s home.
"
While not an outright declaration of their blessing, it is obviously strongly inferred.
Brands
It strikes me as a bit daft for eBay to broadcast their internal findings on this matter.
It doesn't surprise me at all if eBay has found paid ads don't bring much to the table. However that finding is going to be specific to their circumstances I would suspect. With a brand that is so strong, most people willing to deal in a non-retail environment will automatically think of them.
If they can cut the adspend and still make money - they do it. By by brining this up, all they do is state the obvious and potentially piss off Google*.
*And that is just bad for (any) business.
Re: Security by obscurity
In my dealings with Barracuda they have always been forthcoming with the fact they hold their own login points. They are, after-all, a managed-solution appliance provider.
I can't remember the exact wording of their T&C's, but I believe it's in there already.
The fact they had thought to clamp down the IP range in the first place and are now pushing an update to help secure things a bit more is good.
I am not saying that their solution is appropriate for everyone in all fields, but their are many applications where this is perfectly acceptable.
Karl P
Re: Apples and oranges..
There are actually loads of justifications I can think of.
I think the real point though is the one mentioned above.
The current network of observation satellites have no good way of getting data directly to the people who potentially have the most use of it.
Why not just make the current system do/support that? My guess is that they can't. The multi-billlion dollar spy sats that are up there right now seem, more or less, all or nothing affairs that are live all the time and designed to be directed and downlinked only from very limited locales. I would also be surprised if they had any sort of access control built in (AKA, there is a reasonable chance that they _can't_ black out something, aka they can be used to spy on anything that happens to be in their FOV).
In order to open up sat surveillance to a wider group it would seem that we need all sorts of things such as the ability to talk to multiple people simultaneously, to be able to be controlled from one location and video downlinked from many more. to have control over who can access it when, where it needs to black out, be cheap enough that no one is going to overly balk at a wider range of people operating it (no one wants a field grunt of any type to have control over something that costs billions with a b, it just won't happen), etc.
Once you get down into it, it probably makes sense for them to be different systems with different capabilities. If someone wants to say that it also gives the US a leg up on the Chinese sat-killing rockets, (put them up almost as fast as someone else can shoot them down), then all the merrier.
Just a WAG on my part, but one which feels something close to realistic.
Karl P
Re: probably won't notice
I guess it really depends on how you go into it.
If you go in with a bullet point list of things you want to watch, you are going to be disappointed in any of the all-you-can-eat streaming providers (I am excluding Apple / Amazon here as most things are available through them, however it'll cost you dearly for any significant amount of content).
However if you go in expecting quality entertainment without pretext and are willing to experiment a bit, you will struggle to find enough to watch.
I dropped cable and picked up a unthrottled/uncapped commercial internet connection and both hulu/netflix maybe 6 months ago.
I don't watch the same shows anymore (or at least, very many of them) but at last count have over 500 episodes of content waiting to be watched, all of which is very, very good.
A different way of getting content (OTT/streaming/ondemand) also, at least for the moment, necessitates a different strategy for consuming content. If someone isn't prepared to make that leap, then they really shouldn't be dropping their cable/satellite connections.
Karl P
Rackspace and Citrix....?
So you can kill one of the best hosting companies and one of the great remote access/vdi/hypervisor companies all at once?
Seriously, Rackspace is just wonderful as they are, and Citrix is really starting to become exciting again in their own right.
The only positive thing I could begin to think of about this room is "at least it's not Oracle".
Cisco should learn to be competitive again in their core competency. Seriously, these days Cisco is #3 or 4 on my switching provider list and # 5 or 6 on my WLAN provider list. About the only place I look at them seriously is in core routing, but even there I am now trending further rather than closer.
Get me excited about your products again. Make them excellent at a manageable price point. Wrap them in straight-forward services which are both comprehensive and fair.
Do all that and then maybe we could talk about new additions to your company.
I don't find it unreasonable at all.
Some cheap insurance if you ask me.
Any person with a vague understanding of how the modern world works probably has one reason or another to have some sort of plan b.
I can't tell you the last time I used cash for a purchase, much less a major one, but yet I still keep a reasonable amount of Cash handy in the event that I were to ever need it.
In this day and age, when even having the same name as someone else who happened to do a bad thing can be enough to have accounts frozen and services suspended (by error of course, but inconvenient at the very least), I think it might actually be a good idea.
Certainly a much better investment than any number of other dubious "insurance" policies one might be offered in the course of life.
Upon reflection, I may even add a pre-paid to the insurance pile.
As my dad always says "Either be prepared, or hang around someone who is" (:
Re: Samsung - designed to fail.
Saying that Apple is using A9 cores is quite disingenuous is it not?
I could just as truthfully say that they are already using A15's.
In reality both, and neither, of those are true.
http://www.anandtech.com/show/6292/iphone-5-a6-not-a15-custom-core
In my world A15 shows interesting promise for certain computing workloads, but I can't help but feel that Apple did precisely the right thing by incorporating what is good about A9 and A15 in their own package that is laser focused on power efficiency and performance as hand in hand requirements.
My guess is that, at least for a time, others (including amazon) may decide they need the same design latitude in order to drive excellent user experiences.
Re: We think there should be more choice...
If your going to fault Apple for something, I don't know that I would start there. Not being able to set default apps as a general topic would be more appropriate, but then I understand where they are coming from by not doing so.
Maybe google should instead be required to open up (paid, licensed - of course) access to their raw vector data. Particularly seeing as that's why Apple had to leave in the first place.
Re: Gullible hacks...
Because adding cores is only useful if you are running multi-threaded applications (many that you would think to use for casually judging performance aren't) or seriously multi-tasking.
Beyond 2 cores is really only useful to a group that while large in absolutely terms, is relatively quite small.
It's just the continuation of the GHz wars at some level.
@Mad Hacker
You must be walking away from a lot of products then, Almost all of them, in fact.
Practically every streaming project I have worked on in the past two years involves transcoding into a variety of formats and resolutions. The only exception is when we are doing on the fly transcoding, but that is so proc intensive we only use that in special circumstances.
Hopefully that will change in the future. In fact, if we could only have a single format, I'd be happy to deal with the bitrates.
Alas, the world we live in is not (nor ever will be) perfect, and you do what you need to do.
Glad to hear we may begin to get some consensus on an HLS standard and a DRM package to go along with it. Sorely missing at the moment.
Don't think they did anything special for him.
This is just SOP for them.
I have a few acquaintances which have been privy to the exact same treatment for charges even more mundane than those above.
Whether it is excessive or not is besides the point, just wanting to make it clear that you could be Joe Schmo accused of a violation in not having proper paperwork to ship cardboard boxes internationally and would get very similar treatment.
The only difference here is in how publicized the case has become, which may ultimately become its undoing.
I think the title needs to be fixed.
There is no need to further placate the "analysts". If you want to say they missed street numbers, thats fine, but stating analysts guidance numbers as if they are "real" when they have no real insight into the workings of the companies is wrong, potentially even misleading.
Also, when dealing with such a predictable product cycle as Apples, It would seem that YoY numbers are the ones to really pay attention to. In that light I think they did quite well.
There is nothing wrong here, from what I can tell a single availability zone went down.
AWS is designed in such a way that if availability isn't important, you can base your load in one local (in this case, N.Virginia). If you want more availability, use best practices and spread your loads around.
The real story is that these services still aren't properly able to cope with the conditions of the underlying infrastructure.
Posted in I need a New Laptop!!
Macbook/Latitude
Take a look at a 15" macbook and either wipe and bootcamp, or perhaps partition and dual boot. They run win7 absolutely beautifully. Other than some issues with the express card port in the 17" version, they are perfect. I oversee dozens of them at this point.
On the other side, take a look at the dell latitude e6520. You can kit it out with SSD, i7's, up to a fullHD screen, etc.
No cheaper than the MBP however. You can buy a better warranty though..... However you may actually need it more often.....
99p/c/etc?
I am assuming, potentially incorrectly, that the engineers in question are building something much more compelling than a fart app.
If so, 99p/c/etc is a terrible price point.
The only people playing down at that cesspool are doing so willingly.
While I must admit to having a few of these apps on my devices, in general terms the median cost of an app on my iDevice is probably approaching 8$ at this point.
The story here isn't people can't make money, its that people are failing business planning 101.
Quantum Internet
You know what's going on every single one of my RFP's from here on out.
Quantum Internet.
Poor Signaling
Read "Poor Signaling" as in "Vastly more traffic that breaks our usage models". And I doubt they are only talking about Apple here.
While I am general sympathetic to the struggles of large infrastructure companies, this whole thing wreaks a bit to much for me.
Yes there were signaling issues, most of which would have been fixed in phones and base-station firmware updates by this point.
What has remained is an entirely new usage paradigm, absolutely decimating the old long-term growth trajectories.
This necessitates rapid, over-arching capacity buildout everywhere from the radios themselves and their density, to the frequencies allotted and backhaul provisioned.
While the carriers are working on dealing with these issues, they are also tasked with increasing revenue and profits for their shareholders. This has put them in the unenviable position of needing to maintain exceptionally high levels of CAPEX investments in a market which is mostly saturated and a client base which are already near the upper end of what they are willing to pay for the service.
The only place for these funds to come from at this points would be new revenue opportunities (such as selling content or acting as a payment provider) or to decrease costs they already have (say, lower wholesale costs on your popular smartphone de'jure, or maybe decrease the going rate of spectrum?)
What we are seeing here, I think, are vague plays towards all of the above.
Not saying any of it is right or wrong, but I do wish they would just outline what they see the real issues as, and what they would like to see as potential solutions for them.
Then we (as in anyone from governments to individuals, businesses to the telcos themselves) could have a meaningful discussion about the entire matter.
Karl P
I'm quite worried that the reg is going soft....
There was certainly room to take a crack at getting celestial bodies into the headline, but you missed it.
May I request a rewrite?
Thanks, Karl
Posted in Netflix: How to completely screw up
To go all at once, or slowly bleed out?
It's simple really, and the fact so many people are up in arms about it mystifies me.
The management and board of Netflix took a look at the next two to three years and realized that the business wasn't going to be sustainable. Everything from postage rates, to disc acquisition, streaming, storage, and of course digital content acquisition. It's all going up. They had to come up with a new plan.
The plan they decided on (With access to way more information than us side-liners have, it must be noted) was to separate the streaming and physical sides of the business and price them according to their near-term realistic cost structures.
They knew this wasn't going to be popular, and the needed price hikes most certainly wouldn't be so they had a decision to make, either change it all at once - rip off the band-aid so to say, or bleed it over the period of a year or so. First new customers, then multi-dvd, then single dvd subs.
It's pretty obvious to both us and them that the price hikes would cause a reasonable drop in subscribers. Hopefully that would be counteracted by new subscribers, but nonetheless everyone had to know that it would happen.
It's important to stop for a moment and make one fact clear: at this moment Netflix has no valid, legal, competitor. While that may (and hopefully will) change over the next few years, the reality is that right now Netflix stands alone.
That last bit is important because if Netflix goes for the bleed out now, they run the risk of having a competitor for people to switch to later, whereas right now it's more or less an all-or-nothing approach.
Additionally, by finally doing it all at once they all but guaranteed any serious competitor will come in at the now-standard pricing (or even higher), and when doing deals the content owners will be expecting payments based on that cost structure.
Finally, after seeing just how severe the backlash was (more so than they anticipated would be my guess) they decided to go ahead and finish the plan, separate the businesses and start re-building.
They are neither short-sighted or stupid. It's a reasonable plan on their part and hopefully one which will work out well for them long term, albeit with a short term cost.
The pricing is still beyond fair, the selection is reasonable for the price, and they now can spend more time working on streaming content acquisition than physical aquisition.
Karl P
You're are missing the point.
This has nothing to do with giving Tim some huge allotment of money.
This doesn't even have anything to do with a performance bonus (although indirectly it does apply).
As the article and various of you all note, his regular compensation is fair enough already.
No, this is all about succession planning. The board wants to guarantee a stable leadership figure over at least the next 5, preferably 10, years. This is for both public image and shareholder confidence, as well as to bring up and train the next executive team in "the Apple way".
And for that, no amount of money would be too much. The fact that it is a relatively modest amount considering Apples profit, marketshare and "earnings potential" means that there are already other compensation plans underway.
Whether you agree with the Apple ethos or not doesn't matter. The fact remains it is one of the largest companies in the world. Successfully leading such a beast is not only terribly difficult, but deserves compensation commiserate with the responsibilities.
Karl P
If lightsquared are right....
And it's an issue with radio filtering, which would make sense, the gps people are really in the wrong here.
Capable rf filtering is a well understood field, and the only reason transmissions which are that far out of band would be causing issues is if they were beig lazy and or ignoring it altogether.
If the FCC decides to shutdown lightsquared I would hope it's only because a large consortium of GPS manufacturers and users decide to reimburse lightsquared for all costs and further agree to license all bands which their designs fail to filter against heretoforth.
I'm goig to go out on a limb here though and guess between the military, FAA, and private industry sectors that lightsquare is going to get shafted, possibly without am admission of guilt on the gps side.
Karl P
Posted in Apple bars WinXP users from iCloud
The sparks are flying....
At least I think it is sparks because it seems you guys have one hell of an axe to grind against Apple righ now. I can respect if you guys don't agree with all of their decisions, but at the end of the day I can't escape the feeling you guys are being unrealistically hard of late. XP is three revisions back, and the better part of ten years old. If any company starts moving their newest features to a newer OS, I don't think any of us can blame them. Heck, Microsoft is the one who made xp and they are practically done with it. Lighten up the tone guys.
*sigh* - Silly People
Reg: I don't care for the tone of this article. It's nothing but a cheap shot with no thought as to why this may have happened.
Let's look at this logically for a second.
1. It's an AIO. What goes on inside shouldn't concern you much (although how reliable it is as a whole should).
2. If you don't agree with #1 buy a Mac Pro, a linux, or win7 box. The iMac (and probably the portables) will not be for you.
3. Steve Jobs (or an SVP, etc) did not sit around some table musing about ways to screw people out of money and come up with fitting a custom HDD. Thats just silly.
4. What actually happened is engineers made a call that a more accurate HDD temp sensor system made a measurable impact in service life. (thus making the entire unit, which is what we are concerned about in an AIO, more reliable)
5. This is not about money to such a degree that they are actually spending more money to do this. (If it isn't a bog standard drive with bog standard cables/firmware, it is costing them more. Guaranteed.)
6. Someone somewhere along the way brought up the point that this would stop, or at least seriously hamper, customers from being able to pop the case and change the drive. - To that end they would have looked at overall statistics and realized that 90%+ of all users used External HDD's when it was time to upgrade, and also talked to applecare who said opening the machine voided the warranty. (whether they look the other way on a regular basis or not is a different matter).
7. Another reason I wouldn't get too hung up about it; Chances are very good that the rest of the line (With the exception of the MP, where this sort of thing actually matters to an appreciable percent of their customer base) is going this way as well. I anticipate a not-to-distant future where every part of the system is soldered down and compacted as much as is physically (as in, the laws of physics) possible. An upgrade will be changing the entire logic board.
7b (Edit). And you would be remiss if you don't think the general computer marketplace isn't looking at what Apple is doing here. Good bad or indifferent the sales figures and stock numbers are telling a story, and that is that people are willing to spend good money on highly integrated, non serviceable products. (assuming the support is there for when things go wrong). This is the biggest case for Thunderbolt in my mind. It gives us a good, fast, low level external I/O capable of real work. Allowing it to fail may doom us to a period of highly integrated, low or zero expansion systems with no high quality low level external I/O options.
8. Even if you don't agree with anything I said, within 16 months someone will have an elegant hack for it.
Calm down people, it's not that big of a deal. It's actually understandable, and if you look at it from a realistic viewpoint it is probably a reasonable decision.
Karl P
"Main Credit Card Database"?
So.... how long until the announcement of a minor breach of the "Main Credit Card Database".
I wonder if they regret going after the PS3 hacker now....... Nah, they probably don't see these as even remotely related.
And yet....
This isn't a "drive-by" attack, there is outright social engineering involved to get someone to run this app, and as we all know once you can convince a user to run a program, all bets are off.
Furthermore, once it is on a system, to remove it all you do is kill the task and delete the program right there in applications.
I have no doubt that something truly nasty for OSX is coming someday, but this still isn't it.
Posted in US legalizes jailbroken iPhones
Calm down for one second.
Just remember, this is _ONLY_ for the actual legality of the cracking itself. This provides some safety to those writing and hosting the software used for these activities, but on the consumer side of things we still have the actual terms of service agreements and warranty provisions on the side of the carries and manufacturers respectively. Just because it is no longer illegal under the DCMA to crack under these specific circumstances doesn't change the fact that the contracts (can and do) still stipulate you may not unlock the phone (Carrier TOS/Contract) and that if you do you may be subject to actions against yourself (Carrier TOS/Contract) and your warranty on said devices will be gone (Manufacturer Eula/Warranty Provisions).
So while it isn't "illegal" anymore, it remains a clear breach of (legally binding) contract and both the carriers and manufacturers will continue to wield the same axes as before.
In fact I wouldn't be surprised if we see a them sharpening those axes a little more now. From a manufacturers standpoint, "kill switches" like the new Droid has may begin to make more sense and thus become far more common place now that they have lost the "legal" channel over which they could go after the people making the cracking software. If they can't deal with the issue at the high level, it may make sense to enforce it at a more general, user level.
I hope I am wrong, but I just can't help but feel the EFF has once again done a single right that will result in a greater wrong.
Karl P
Is Required
While not a fanboy, I generally understand Apple's position from a business aspect, and from that angle I can understand why the no execution thing is in place.
In this particular case however I have to agree that Opera is on to something. What they are doing is not appreciably different than numerous apps that are already in the store.
While the Apple brand can afford a lot of the press that presents itself from some of the app store opportunities that have/will arise, it would be incredibly short sighted for them to deny this app.
However, if it is anything like mini on the numerous other platforms I have used, I don't believe it will gain much traction, it just isn't that good of an app in comparison to TouchOS-Safari. Maybe the new V is substantially better?
